Design of a 3-axis robot arm and classification of EEG data with machine learning to control the robot arm

It is thought that in the future, human-machine interaction will be widely achieved using thoughts and brain signals. With the discovery and spread of new application methods on brain-machine interface, it is expected that a new period of development will be experienced in technology, daily life and industrial production sector. Research and studies on the subject will contribute to the literature and will be useful in terms of accelerating the process and catching up with the developing technology. For this purpose, within the scope of this Master's Thesis, a 3-axis version of the robot arms widely used in the industry was created using PLC circuit, and for its control, the data obtained by EEG method was classified with machine learning. With the use of brain-machine interface on such machines it is aimed to increase the practicality and efficiency of control. In addition, it will increase the suitability of industrial machines for use by physically disabled individuals. Another aim of the project is to create a basic information source on the design and production of a mechatronic system. Firstly the mechatronic system consisting of the mechanics, electronics and software of the robot arm was designed and put into production. Secondly the brain signals representing the researcher's direction commands were recorded with an EEG device, classified with machine learning algorithms necessary for the manipulation of the robot arm. Testing of the robot arm is done by moving a suitable object to a different point within the workspace. As a result of the studies, it has been concluded that control applications with brain signals offer an alternative to current control applications, but for active use in the sector, better technology and methods must first be developed in the field of brain-machine interface.